# ai-research-reproduction

> Claude/Codex skill by [@lllllllama](https://omgskills.com/library/lllllllama/index.md) · 484 stars · 357k installs · Updated 2026-09-23

Rigor Reproduce compatible skill slug for README-first deep learning repository reproduction. Use when the user wants an end-to-end, minimal-trustworthy flow that reads the repository first, selects the smallest documented inference or evaluation target, coordinates intake, setup, trusted execution, optional trusted training, optional repository analysis, and optional paper-gap resolution, enforces conservative patch rules, records evidence assumptions deviations and human decision points, and writes the standardized \`repro\_outputs/\` bundle. Do not use for paper summary, generic environment setup, isolated repo scanning, standalone command execution, silent protocol changes, score chasing, or broad research assistance outside repository-grounded reproduction.

Tags: agent-skills, ai-agents, deep-learning, reproducibility, research-tools

## Install

```sh
git clone https://github.com/lllllllama/RigorPilot-Skills /tmp/RigorPilot-Skills && ln -s /tmp/RigorPilot-Skills/skills/ai-research-reproduction ~/.claude/skills/ai-research-reproduction
```

## From README

ai-research-reproduction Purpose Guide README-first deep learning reproduction toward the smallest trustworthy run with auditable evidence. Preserve documented meaning; record assumptions, deviations and blockers instead of changing semantics to manufacture success. Load specialized references only for a concrete uncertainty. Fast Path For a routine bounded run, keep the control path short: Read the target README and only the target test/config/source needed to understand the documented command. Run scripts/orchestraterepro.py --repo --plan-only --agent-output with any explicit user timeout bound (--timeout or --train-timeout) already supplied; review commandcandidates, the selected cmd-XX, side-effect contract, selection fingerprint, and returned reviewedrunargs. With no --output-dir, later evidence goes to /reprooutputs regardless of caller cwd.

## Links

- HTML page: https://omgskills.com/library/lllllllama/rigorpilot-skills/ai-research-reproduction/
- GitHub: https://github.com/lllllllama/RigorPilot-Skills
- Author: [@lllllllama](https://omgskills.com/library/lllllllama/index.md)

## More from this repo

- [paper-context-resolver](https://omgskills.com/library/lllllllama/rigorpilot-skills/paper-context-resolver/index.md) — Rigor Paper Context helper for README-first deep learning repo reproduction.
   - [@lllllllama](https://omgskills.com/library/lllllllama/index.md) · 516 stars
- [repo-intake-and-plan](https://omgskills.com/library/lllllllama/rigorpilot-skills/repo-intake-and-plan/index.md) — Rigor Intake helper for README-first deep learning repo reproduction.
   - [@lllllllama](https://omgskills.com/library/lllllllama/index.md) · 516 stars
- [env-and-assets-bootstrap](https://omgskills.com/library/lllllllama/rigorpilot-skills/env-and-assets-bootstrap/index.md) — Rigor Setup skill for README-first deep learning repo reproduction.
   - [@lllllllama](https://omgskills.com/library/lllllllama/index.md) · 486 stars